GFCI circuit for pool lights trip when turned OFF ???

[ritelec]

Good morning all.

I installed three low voltage pool lights (controlled from 3 way switches).
All seemed to be working well.
Got a call back that the breaker is tripping. So. Here it is.

3- lights..
3- deck boxes..
3- 120v to 14v 70va xformers.

The owner reported the tripping and that he was having a tough time setting the color of one of the lights (they cycle through color changes).

I removed the xformer from the circ. for the light he mentioned. the other lights would turn on and off, all seemed fine. Then to clarify if it was if it was the xformer or light I took the "known"
good xformer and put it at that deck box......then after cycling the switches on/off several times...when turned off the breaker
would trip.
I removed the load from 2 of the transformers and removed 2 transformers individually and swapped them and..... the circ would hold than after turning the switch on and off "several"
times it would trip again when turned off. Sometimes after turning them on and off "ones".
One time I went to reset the breaker and it tripped as I reset the breaker with lights off.

I'm thinking the travelers of the 3 ways may be damaged but tripping would be happening when the damaged conductor is energized. And with the 3 ways being in different states
for the most part the lights come on all the time and cycle on and off till turned off that last time when it trips.

So- I'm thinking bad light? bad xformer? nicked conductor? garden gnome?

In trouble shooting, I wound up temping a drop light (120v/60w A lamp) to the lines of one of the deck boxes so I didn't have to keep going back and forth to look into the pool to see if the
lights were on or off.

With the A lamp hooked up, and with the xformers and lights hooked up or not hooked up, the lights in the pool turned on and off and worked every time. 50 or more on and off's from either 3-way on and off...
No problems. BUT.... When I removed the A lamp from the circuit, the lights would trip again.

Currently... I hooked everything back up the way it was and the way it "should be"??.
The 3 ways control a lighting contactor in a control panel so at that location I spliced in the 60w bulb.
All is working fine and the way it is supposed to... lol.. the only issue is everytime the owner turns on his pool lights, a light turns on at the other side of the yard... :- )


Any thoughts why these x formers are tripping the circ when turned "off"?
If I could have assistance with my math...... I believe the fixtures themselves were slightly too large to have two lights on one x former so I'm thinking the fixtures are 40w each.
What would be the load for these three xformer/fixtures on the 120v side?
At the 14v side, would va be based on the 70w x former or the 40w fixture it supplies.
If at 70va .... would the 5 amps transpose proportionately to the 120v side or is 5 amps 5 amps? (been a while)
I can only guess that the load is milli-ampish so the breaker is having nuisance tripping, but putting a real load (like the 60w lamp) is holding the circuit...????
Might this be correct?

My best thought for getting through this (and not have the light at the other end of the yard come on with the pool lights) is to install a resistor at the pool control panel
location. 120v resistor??? What color stripes, what tolerance???


Never saw this before. Any thoughts?
Thank you,
Rich

 

[gar]

130720-1027 EDT

If I follow what you have said you get an inductive kick when you turn off the transformer to the low voltage lights, and the GFCI is triggered off by this large voltage spike. I believe your test lamp load, basically a resistor, was shunting the transformer primary and limiting the transient spike. But I really did not follow your description.

If it is a turnoff switching transient, then an MOV, capacitor, resistor, or light bulb may be the easy solution.

.

 

[ritelec]

Thanks Gar,
Sorry for the discription, but yes I think you got the jist of it.

I was doing the on/off thing and as I was doing it I was thinking of what you mentioned. Like energizing an xformer with a battery. Removing the battery and then the field collapsing.

If this is the case..............do others installing low voltage lighting run into this problem?

Why would they make these fixtures and low voltage transformers if this is what happens?

Does it not happen on some install and happen on others???

Nutty.

Metal oxide varistar ????? what the.......... :- )

Rich

 

[Little Bill]

And with the 3 ways being in different states
for the most part the lights come on all the time and cycle on and off till turned off that last time when it trips.

That right there is your problem. 3-ways and GFCIs don't work well when in different states, especially in north eastern states!:lol:

Also a huge voltage drop when ran that far!:angel:

 

[gar]

130720-1331 EDT

ritelec:

See my posts 3 and 9 at:
http://forums.mikeholt.com/showthread.php?t=152937

Whether it was rate of rise, peak voltage, or a combination that resulted in false tripping of the GFCI I don't know. By putting something across the load that reduces either or both factors and results in elimination of the false trip, then this can be considered a useful solution without trying to find a less susceptible GFCI.

A lamp wastes energy continuously when the circuit is on, a capacitor does not, and an MOV has very little dissipation at normal line voltage. An MOV does not eliminate initial rate of rise, but clamps voltage over a range lessening total change of voltage. A capacitor slows down rate of rise, and reduces peak voltage from an input impulse. Impulse is a very short transient of some energy level.

I do not know how different GFCI brands compare in sensitivity to false triggering.

.

 

[ritelec]

Thanks Gar.

Nice catch Little Bill...Very funny :- )

"A lamp wastes energy continuously when the circuit is on".......
you mentioned MOV and capacitor, you also mentioned resistor originally.

I wouldn't know how to size for any of the above, but I would think I have a better shot looking for a resistor.....???

Or as mentioned, look for a better GFCI. the intermatic panel is rated for several manufactures.... I guess I could always ask my supplier for a handful????



thank you
Rich

 

[beanland]

Defective Transformer

Defective Transformer

How about this: the transformer that is giving you grief has weak dielectric. When you turn it off, the field collapses and generates a voltage spike. That spike is enough to short to ground. That short to ground is what the GFCI is seeing and it trips. Can you megger the transformers without damaging them? BTW, are your transformers real iron core units or modern solid state wall warts? That might make a difference.

 

[gar]

130720-2554 EDT


Try a Vishay VDRS14t1150BSE page 697 of Mouser, about $0.80 plus shipping and any minimum.

This has a maximum continuous AC RMS rating of 150 V, 1 mA at 240 V DC, 400 V clamp at 50 A, max energy 50 Joules, and maximum non-repetitive 4500 A, but probably not exceeding the Joule rating.

.

 

[ritelec]

How about this: the transformer that is giving you grief has weak dielectric. When you turn it off, the field collapses and generates a voltage spike. That spike is enough to short to ground. That short to ground is what the GFCI is seeing and it trips. Can you megger the transformers without damaging them? BTW, are your transformers real iron core units or modern solid state wall warts? That might make a difference.

Iron core or solid state??? Iron core ....I believe. Looks like a transformer, larger and heftier doorbell or boiler type mounted in a hand plastic support that slides into a cover that
attaches to the plastic deck box.

I don't believe it's a one of the transformers .

I first removed the questionable transformer ..... the circuit held.....

I then took another transformer that was "right" next to it (for a different light) to confirm the light was ok.
It tripped. Removed the load (light) and it tripped again. (thought 12" made a difference...why??? and scratching my head).

Flopping xformers back and forth on those two deck boxes, hooking lights up and removing.............the circuit did not hold.
Nothing held the circuit till I added the bulb.

Gonna check out the suggestion Gar thanks.

Also use it or loose it..........

E=IxR
120=.5xR
whats R..............my my my.

Been a long time................ 240 ohms???

Was also wondering what the ohms would "have" to be to hold.

Was wondering about inserting a 7w night light to see if it would hold then get the right resistor.

 

[ritelec]

I gotta look again what you are referring to Gar

Mouser catalog 646 page 697...no mention that I could see of the Vishay..

I see info for On Semiconductor, cdm protection..transient protection.......they look like little circuit boards???

 

[eHunter]

I gotta look again what you are referring to Gar

Mouser catalog 646 page 697...no mention that I could see of the Vishay..

I see info for On Semiconductor, cdm protection..transient protection.......they look like little circuit boards???

Try this URL:

http://www.mouser.com/ProductDetail...D1w7EfmGkhMNVFdT92%2bbW0QVtslEJFayTF9yY7wDA==

 

[hurk27]

If you can use a Square D home line they have a little more resistance to inductive voltage spikes, I used to think it was because of the high frequency nature of a transient spike that was causing an out of phase spike but as another member pointed out that the current from the spike is on both the hot and neutral so it wasn't the problem, so now I feel it is a problem of capacitive coupling with other conductors or EGC path that is why transients are causing this problem.

I have played with MOVs in several cases of having these problems when I had an apartment that had cheap GFCI receptacles that were tripping almost every time an inductive load was switched off, I tried putting the MOV line to neutral at the load, line to neutral at the switch, but the transient spike was just to fast from the MOV to react, messing around I put the MOV across the switch and to my surprise it almost totally went away, it would still do it but it took forever to get it to do it again, so I figured that by placing it across the switch it allowed the surge to bleed off back through the load, then I tried a bi- poler capacitor thinking of the caps older vehicles use to lower damage to the points in in the distributor from the inductive kick back from the coil, and I could never get the GFCI's to trip again, I didn't leave the cap in place because I just changed out the cheap GFCI's with some Leviton which were less susceptible to this problem but it did work, it might take some experimenting to find the correct value as I had a bunch of them of different values and I have no idea which one I used to get it to stop but gar might be able to shed some light on where to start, also once the cap charged unlike a resistor it would not pull much in the way of power the reason I didn't leave it this way was because I was worried that when you turn on the switch it would short out the cap causing more contact wear on the switch, but like I said since you have the option of using a different manufacture for the breaker I would try going that route first instead of Mcgyvering a solution.

 

[mgookin]

delete

 

[gar]

130721-1142 EDT

hurk27:

I believe the MOV was most effective when across the switch because it clamped the peak voltage at the circuit breaking point. When the MOV is across the load it only clamps the voltage at the load. There is additional inductance in the circuit, at least the supply transformer.

To use a capacitor as a snubber it usually has a resistor in series to limit the peak inrush current. A capacitor with zero initial charge (zero volts across the capacitor) looks like a short circuit when connected to a source of energy.

.

 

[ritelec]

Thanks guys...............

So I have this situation where 3 transformers controlled from a pair 3 way switches trips the gfci circuit breaker when the switch is turned off.
Any thoughts???

LOL

Not knowing here but, upon the readings... the MOV is poke and hope and questionable.
Another breaker manufacturer may solve the issue (currently there is a Murray being used, I could also try a Siemens but aren't they the same as Murray?
I believe there may have been two maybe three other manufactures listed, but looking high and low Intermatic gives all kinds of info on their panels, but does not list their approved breakers or
show you the sticker with the approved brks that is on their door)

I guess I could go there with the several approved breakers, but it's not a definite repair right? I mean they could all trip?
Not to mention that if one works, I'm not sure the supply house would take the return of the others (that's if one worked) for a test.

I'm thinking RESISTOR...........

Don't play with it to see if a 7 w or a 40 w or...... would work.
60w A lamp holds (I guess, I haven't heard anything)...... stick a resistor in and hopefully be done with it.

Do Resistors have a life expectancy ????
I guess everything breaks or dies.................


Thanks again gents..
Rich

 

[gar]

130721-1331 EDT

ritelec:

A wild guess is that an Ohmite 25 W wirewound 10 ohm resistor operated at 5 W in free air at 20 C air temperature and no mechanical damage will probably last thousands of years. If not more.

.

 

[ritelec]

130721-1331 EDT

ritelec:

A wild guess is that an Ohmite 25 W wirewound 10 ohm resistor operated at 5 W in free air at 20 C air temperature and no mechanical damage will probably last thousands of years. If not more.

.

thousands of years if not more.................ok, and then what, I really don't want to have to do another call back..........LOL

Thanks Gar....

Question.... is that what I should go for in my case, or is that a hypothetical resistor that will last a thousand yrs.?


Do I have it right ????

E=IxR I= 25w/120 I=.208
120= .208xR
R= 24.96 ohms

I=IxR
120=Ix10
I= 12

Sorry, not getting it, where am I off???

What should I get? A 60w 240ohm?

Thanks

 

[eHunter]

thousands of years if not more.................ok, and then what, I really don't want to have to do another call back..........LOL

Thanks Gar....

Question.... is that what I should go for in my case, or is that a hypothetical resistor that will last a thousand yrs.?


Do I have it right ????

E=IxR I= 25w/120 I=.208
120= .208xR
R= 24.96 ohms

I=IxR
120=Ix10
I= 12

Sorry, not getting it, where am I off???

What should I get? A 60w 240ohm?

Thanks

Voltage (E) of 120 is known
Standard Resistance (R) of 2.7K Ohm is selected
I = E/R = 120/2700 = 0.04444
P = E*I = 120 * 0.04444 = 5.3333

Voltage (E) = 120 Volts
Current (I) = 0.04444 Amps
Resistance (R) = 2700 Ohm
Power (P) = 5.3333 Watts

There are several ohms law calculators available online that can possibly help you

https://www.google.com/search?num=1...din,hmss2=false...0...1.1.21.serp.fVU7gOo1Wx4

https://www.google.com/search?num=1...in,hmss2=false.1..0...1.1.21.serp.UVrwqgiRByE

25 watt, 2.7K Ohm resistors

https://www.google.com/search?num=1...din,hmss2=false...0...1.1.21.serp.utkdY2YWK8o

 

[cadpoint]

If you took off all the devices, use a meggar on some wires.

 

[gar]

130721-1944 EDT

ritelec:

If you really want to minimize the problem with a resistor, then use two 120 V incandescent bulbs in series connected to the circuit about where you connected your test light. If your test light was a 100 W, then I would use 150 W for each of the series lights. At about 50% voltage on each bulb the life time will be very long, possibly 100 years if there is no abnormal condition over that time and the bulbs were well sealed. However, on-off cycling might shorten the life.

At 50% of nominal voltage applied to a tungsten filament bulb the input power is about 1/3 of rating. Thus, two 150 W 120 V bulbs in series when powered from 120 V will require about 100 W.

Further this should allow you to use all UL components, and conventional wiring.

.